One of the major interests in the field of Second Language Acquisition (SLA) is to uncover the mechanisms involved in learning a second language in adulthood. Although adult learners’cognitive system is more well-developed than that of children acquiring their first languages, children are much more capable of perfecting their language than adults acquiring a second language.
The current study focuses on the acquisition of overt pronouns by English-speaking JFL (Japanese as a Foreign Language) learners. Japanese has both covert (null) pronouns and overt pronouns, while English does not have covert pronouns in tensed clauses. Overt pronouns in Japanese and English differ in that English overt pronouns allow both coreferential and bound variable readings while in Japanese, the so-called overt pronouns kare and kanozyo can have coreferential reading, but do not allow bound variable reading. However, covert pronouns can have both coreferential and bound variable readings. This was explained by the Overt Pronoun Constraint, the principle which was proposed to be one of the properties of Universal Grammar (UG). By looking at how American JFL learners interpret Japanese overt pronouns, we would know whether the OPC is at work. If the OPC is part of UG and if their grammar is constrained by UG even during the development of the target grammar, they would not permit Japanese overt pronouns to be bound variables.
The current study attempts to answer the following three research questions: (1) Are the coreferential readings of Japanese overt pronouns acquired by L2 learners? (2a) Do they have the knowledge that the bound variable reading is not available with Japanese overt pronouns? And (2b) if they do, when is this knowledge acquired?
The experiment was designed to answer these questions by employing a truth value judgment (TVJ) task. Sixty-one JFL learners in four different levels and 20 native speakers participated in the study. Four types of test stimuli investigated whether the participants can correctly accept coreferential member and coreferential non-member readings, and whether they can correctly reject the bound variable readings of overt pronouns.
The results showed that the JFL learners have firm understanding of the coreferential reading of Japanese overt pronouns, but their correct response rate was clearly lower in the sentences with bound variable readings. It is unlikely that the JFL learners are capable of applying the OPC through access to UG, assuming that the OPC is true. The learners’ correct response rates in the bound variable reading sentences showed a gradual increase as the learners’ language proficiency level went up. The acquisition of anaphoric expressions takes time in L2 Japanese. With independent evidence, this study provides an additional piece of empirical evidence that the OPC does not seem to hold as a constraint in UG.